- Email: [email protected]
Assessment of Hypoxia in Human Cervical Carcinoma Xenografts by Dynamic Contrast-Enhanced Magnetic Resonance Imaging
Int. J. Radiation Oncology Biol. Phys., Vol. 73, No. 3, pp. 838–845, 2009 Copyright Ó 2009 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/09/$–see front matter
doi:10.1016/j.ijrobp.2008.10.062
BIOLOGY CONTRIBUTION
ASSESSMENT OF HYPOXIA IN HUMAN CERVICAL CARCINOMA XENOGRAFTS BY DYNAMIC CONTRAST-ENHANCED MAGNETIC RESONANCE IMAGING CHRISTINE ELLINGSEN, M.SC., TORMOD A. M. EGELAND, M.SC., KRISTINE GULLIKSRUD, M.SC., JON-VIDAR GAUSTAD, M.SC., BERIT MATHIESEN, B.SC., AND EINAR K. ROFSTAD, PH.D. Group of Radiation Biology and Tumor Physiology, Department of Radiation Biology, Institute for Cancer Research, Norwegian Radium Hospital, Montebello, Oslo, Norway Purpose: Patients with advanced cervical cancer and highly hypoxic primary tumors show increased frequency of locoregional treatment failure and poor disease-free and overall survival rates. The potential usefulness of gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA)–based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in assessing tumor hypoxia noninvasively was investigated in the present preclinical study. Methods and Materials: CK-160 and TS-415 human cervical carcinoma xenografts transplanted intramuscularly (i.m.) or subcutaneously (s.c.) in BALB/c nu/nu mice were subjected to DCE-MRI and measurement of fraction of radiobiologically hypoxic cells. Tumor images of Ktrans (the volume transfer constant of Gd-DTPA) and ve (the extracellular volume fraction of the imaged tissue) were produced by pharmacokinetic analysis of the DCE-MRI data. Fraction of radiobiologically hypoxic cells was measured by using the paired survival curve method. Results: Fraction of radiobiologically hypoxic cells differed significantly among the four tumor groups. The mean values ± SE were determined to be 44% ± 7% (i.m. CK-160), 77% ± 10% (s.c. CK-160), 23% ± 5% (i.m. TS-415), and 52% ± 6% (s.c. TS-415). The four tumor groups differed significantly also in Ktrans, and there was an unambiguous inverse relationship between Ktrans and fraction of radiobiologically hypoxic cells. On the other hand, significant differences among the groups in ve could not be detected. Conclusions: The study supports the clinical development of DCE-MRI as a method for assessing the extent of hypoxia in carcinoma of the cervix. Ó 2009 Elsevier Inc. Cervical carcinoma, DCE-MRI, Hypoxia, Radiation sensitivity, Xenografts.
Preclinical and clinical studies have suggested that gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA)–based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may be a useful noninvasive method for characterizing the physiologic microenvironment of tumors (11). A DCE-MRI–based method is particularly attractive in cervical carcinoma patients because DCE-MRI is an established and commonly used method for depicting the primary tumor in these patients (12). Moreover, initial investigations of the prognostic value of DCE-MRI in carcinoma of the cervix have given promising results. Mayr et al (13, 14) observed that a high relative signal intensity in the primary tumor was associated with a high local control rate in patients given radiation therapy. High levels of contrast enhancement could be a result of efficient blood perfusion and thus reflect good oxygenation and hence increased radioresponsiveness. Alternatively, high levels of contrast enhancement could be a result of large extracellular volume fractions and thus reflect low numbers of clonogenic cells and hence increased radiocurability.
INTRODUCTION Advanced squamous cell carcinoma of the uterine cervix is primarily treated with radiation alone or radiation in combination with surgery and/or chemotherapy (1). In the 1960s, hypoxia was identified as a characteristic feature of cervical carcinoma that could cause resistance to radiation therapy (2). Recent studies have confirmed these initial observations (3–6) and have also shown that hypoxia may be an adverse prognostic factor in patients given surgery as primary treatment (3). Moreover, extensive hypoxia in the primary tumor has been shown to be associated with biologic aggressiveness, invasive growth, and increased metastatic propensity (7, 8). Thus, cervical carcinoma patients with highly hypoxic tumors show increased frequency of locoregional treatment failure and poor disease-free and overall survival rates (9, 10), and these patients may therefore benefit from particularly aggressive treatment. Consequently, a noninvasive diagnostic method for assessing the extent of hypoxia in cervical tumors is needed.
Acknowledgments—This study was supported by the Norwegian Cancer Society. Received Aug 11, 2008, and in revised form Oct 3, 2008. Accepted for publication Oct 8, 2008.
Reprint requests to: Einar K. Rofstad, Ph.D., Department of Radiation Biology, Institute for Cancer Research, Norwegian Radium Hospital, Montebello, N-0310 Oslo, Norway. Tel: 47-2278-1206; Fax: 47-2278-1207; E-mail: [email protected] Conflict of interest: none. 838
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
There has been an increasing interest in using DCE-MRI for characterizing the physiologic microenvironment of cervical tumors (15–19). Correlations have been found between DCE-MRI–derived parameters and microvascular density, amount of interstitial fibrous tissue, tumor cell density, and extracellular volume fraction as assessed by histologic examinations of the imaged tissue (15–17) as well as tumor oxygenation as assessed by polarographic measurements of tissue pO2 (17–19). However, the correlations are generally weak, perhaps because unsuitable parameters were derived from the DCE-MRI series (i.e., parameters calculated from signal intensity curves without the use of pharmacokinetic models or by the use of inadequate models) or because the DCE-MRI was performed under suboptimal conditions (i.e., inadequate temporal and spatial resolution, low signalto-noise ratio, and/or significant motion artifacts). High-quality preclinical studies are therefore needed to establish whether DCE-MRI has the potential to provide clinically useful surrogate parameters for the extent of hypoxia in cervical carcinoma. The potential usefulness of Gd-DTPA-based DCE-MRI in assessing fraction of hypoxic cells in human tumor xenografts is currently being evaluated in our laboratory. Our studies are based on the hypothesis that tumor hypoxia is the result of an imbalance between oxygen supply and oxygen consumption (20). The oxygen supply is determined primarily by the blood perfusion, and the oxygen consumption is determined primarily by the respiratory activity of the tissue and, hence, the cell density (21). Hypoxic tissue is therefore expected to be found in tumor regions with poor blood perfusion and/or low extracellular volume fraction. We have already shown that DCE-MRI–derived parametric images may provide information on the extent of hypoxia in human melanoma xenografts (22, 23). The purpose of the work reported here was to investigate the feasibility of DCE-MRI in assessing the extent of hypoxia in human cervical carcinoma xenografts. Tumors of two xenograft lines differing significantly in histologic appearance, cellular radiation sensitivity, and fraction of radiobiologically hypoxic cells were included in the study. METHODS AND MATERIALS Mice and tumors CK-160 and TS-415 human cervical carcinoma xenografts growing in adult (8-12 weeks of age) female BALB/c nu/nu mice were used as experimental tumor models. Tumors were initiated from established cell lines cultured in RPMI-1640 (25 mmol/l HEPES and L-glutamine) supplemented with 13% bovine calf serum, 250 mg/l penicillin, and 50 mg/l streptomycin. The CK-160 line was established from a pelvic lymph node metastasis of a 65-year-old woman who had developed a highly invasive, well-differentiated (histologic Grade I), keratinizing primary tumor. The TS-415 line was derived from a pelvic lymph node metastasis of a 45-year-old woman who had developed a highly invasive, poorly differentiated (histologic Grade III), nonkeratinizing primary tumor. The mice were kept under specific pathogen-free conditions and were given sterilized food and tap water ad libitum. Approximately
839
5 105 cells in 10 ml of Hanks’ balanced salt solution were inoculated in the gastrocnemius muscle to produce intramuscular (i.m.) tumors or in the subcutis far back on the flank to produce subcutaneous (s.c.) tumors. Experiments were initiated when the tumors had grown to a volume of 200 to 600 mm3. Tumor volume (V) was calculated as V = p/6 a b2, where a is the longer and b is the shorter of two perpendicular diameters. Both DCE-MRI and tumor irradiation were carried out with anesthetized mice. Fentanyl citrate (Janssen Pharmaceutica, Beerse, Belgium), fluanisone (Janssen Pharmaceutica), and midazolam (Hoffmann-La Roche, Basel, Switzerland) were administered i.p. in doses of 0.63 mg/kg, 20 mg/kg, and 10 mg/kg, respectively. Animal care and experimental procedures were approved by the Institutional Committee on Research Animal Care and were performed in accordance with the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY).
DCE-MRI The DCE-MRI technique was carried out on 15 i.m. CK-160 tumors, 16 s.c. CK-160 tumors, 15 i.m. TS-415 tumors, and 14 s.c. TS-415 tumors by using a procedure described in detail previously (23). Briefly, a 24-G neoflon connected to a syringe by a polyethylene tubing was inserted in the tail vein of tumor-bearing mice, and GdDTPA (Schering, Berlin, Germany) was administered in a bolus dose of 5.0 ml/kg (0.3 mmol/kg) after the mice had been positioned in the magnet. Imaging was performed at a spatial resolution of 0.23 0.47 2.0 mm3 and a time resolution of 14 s by using a 1.5-T whole-body scanner (Signa, General Electric, Milwaukee, WI) and a 13-cm-long, cylindrical, slotted tube resonator transceiver mouse coil with a diameter of 40 mm. Two calibration tubes, one with 0.5 mmol/l Gd-DTPA in 0.9% saline and the other with 0.9% saline only, were placed adjacent to the mice in the coil. The tumors were imaged axially in a single section through the tumor center by using a scan thickness of 2 mm, a number of excitations of 1, an image matrix of 256 64, and a field of view of 6 3 cm2. Two types of spoiled gradient recalled images were recorded: proton density images with repetition time TR = 900 ms, echo time TE = 3.2 ms, and flip angle a = 20 , and T1-weighted images with TR = 200 ms, TE = 3.2 ms, and a = 80 . Two proton density images and three T1-weighted images were acquired before Gd-DTPA was administered, and T1-weighted images were recorded for 15 min after the administration of Gd-DTPA. The Gd-DTPA concentrations were calculated from signal intensities as described by Hittmair et al (24), and the DCE-MRI series were analyzed on a voxel-by-voxel basis by using the arterial input function established by Benjaminsen et al (25) and the pharmacokinetic model developed by Tofts et al (26). Two parameters were determined for each voxel: Ktrans, the volume transfer constant of Gd-DTPA, and ve, the extracellular volume fraction of the imaged tissue. Images of Ktrans and ve were generated by using the SigmaPlot software (SPSS Inc., Chicago, IL).
Fraction of radiobiologically hypoxic cells Intramuscular CK-160 tumors, s.c. CK-160 tumors, i.m. TS-415 tumors, and s.c. TS-415 tumors were irradiated at a dose rate of 5.1 Gy/min by using a Siemens Stabilipan X-ray unit, operated at 220 kV, 19-20 mA, and with 0.5-mm copper filtration (27). Hypoxic tumors were obtained by occluding the blood supply with a clamp 5 min before irradiation. Tumor cell survival was measured in vitro by using a plastic surface colony assay (28). Briefly, the tumors were resected
I. J. Radiation Oncology d Biology d Physics
840
immediately after irradiation, minced in cold phosphate-buffered saline, and treated with an enzyme solution consisting of 0.2% collagenase, 0.05% pronase, and 0.02% DNase (2 h at 37 C). Trypan blue–negative cells were plated in 25-cm2 tissue culture flasks and incubated at 37 C for 14 days. Cells giving rise to colonies >50 cells were scored as clonogenic. The cell surviving fraction of an irradiated tumor was calculated from the plating efficiency of the cells of the tumor and the mean plating efficiency of the cells of six untreated control tumors. Cell survival curves were established for clamped and unclamped tumors, and the fraction of radiobiologically hypoxic cells was calculated from the vertical displacement of the curves (27).
Statistical analysis Experimental data are presented as arithmetic mean SE unless otherwise stated. Statistical comparisons of data were carried out by using the Student t test (single comparisons) or by one-way analysis of variance (multiple comparisons) followed by the Bonferroni test when the data complied with the conditions of normality and equal variance. Under other conditions, comparisons were carried out by nonparametric analysis using the Mann-Whitney rank-sum test (single comparisons) or the Kruskal-Wallis one-way analysis of variance on ranks (multiple comparisons) followed by Dunn’s test. Probability values of p < 0.05, determined from two-sided tests,
Volume 73, Number 3, 2009
were considered significant. The statistical analysis was carried out by using the SigmaStat statistical software (SPSS Inc., Chicago, IL).
RESULTS Histologic findings of the xenografted tumors were similar to those of the donor patients’ tumors. The CK-160 cells gave rise to well-differentiated tumors with a histologic appearance consistent with Grade 1 squamous cell cervical carcinoma (Fig. 1a), whereas the TS-415 cells gave rise to poorly differentiated tumors consistent with Grade 3 squamous cell cervical carcinoma (Fig. 1b). The DCE-MRI data of a representative tumor of each of the four groups are presented in Fig. 2, which shows the Ktrans image, the Ktrans frequency distribution, the ve image, and the ve frequency distribution of an i.m. CK-160 tumor (Fig. 2a), a s.c. CK-160 tumor (Fig. 2b), an i.m. TS-415 tumor (Fig. 2c), and a s.c. TS-415 tumor (Fig. 2d). Significant intratumor heterogeneity in Ktrans was observed, with the highest values in the periphery and the lowest values in central regions. The tumors were also highly heterogeneous in ve and showed regions with low or high values both in the periphery and in the center.
Fig. 1. Histologic appearance of an intramuscular (i.m.) CK-160 tumor (a) and an i.m. TS-415 tumor (b). Sections were stained with hematoxylin.
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
841
Fig. 2. Ktrans image, Ktrans frequency distribution, ve image, and ve frequency distribution of an i.m. CK-160 tumor (a), a s.c. CK-160 tumor (b), an i.m. TS-415 tumor (c), and a s.c. TS-415 tumor (d). (For discussion, see text).
Median Ktrans and median ve of all tumors subjected to DCE-MRI are summarized in Fig. 3. The TS-415 tumors showed higher Ktrans values than the CK-160 tumors, regardless of whether the tumors were transplanted i.m. (p < 0.0001) or s.c. (p = 0.0057). In addition, Ktrans was higher for the i.m. tumors than for the s.c. tumors, both for the CK-160 (p = 0.017) and the TS-415 (p < 0.0001) line. On the other hand, ve did not differ significantly between CK-160 and TS-415 tumors (p > 0.05, both for i.m. and s.c. tumors) or between i.m. and s.c. tumors (p > 0.05, both for CK-160 and TS-415 tumors). Cell survival curves for clamped and unclamped tumors irradiated in vivo and assayed in vitro are presented in Fig. 4. D0 was higher for the TS-415 tumors than for the
CK-160 tumors by a factor of 1.5 (p < 0.0001), but did not differ between i.m. and s.c. (p > 0.05) tumors or between clamped and unclamped (p > 0.05) tumors of the same line. The numerical values of D0 (mean SE) were 3.02 0.09 Gy (i.m. CK-160), 2.97 0.12 Gy (s.c. CK160), 4.50 0.11 Gy (i.m. TS-415), and 4.45 0.10 Gy (s.c. TS-415) for clamped tumors and 2.95 0.13 Gy (i.m. CK-160), 3.04 0.12 Gy (s.c. CK-160), 4.54 0.14 Gy (i.m. TS-415), and 4.49 0.14 Gy (s.c. TS-415) for unclamped tumors. Fraction of radiobiologically hypoxic cells (mean SE) was calculated from these survival curves to be 44% 7% (i.m. CK-160), 77% 10% (s.c. CK-160), 23% 5% (i.m. TS-415), and 52% 6% (s.c. TS-415). The CK-160 tumors showed higher hypoxic
842
I. J. Radiation Oncology d Biology d Physics
Volume 73, Number 3, 2009
DISCUSSION
Fig. 3. Median Ktrans (a) and median ve (b) of 15 intramuscular (i.m.) CK-160 tumors, 16 subcutaneous (s.c.) CK-160 tumors, 15 i.m. TS-415 tumors, and 14 s.c. TS-415 tumors (for discussion, see text). Symbols represent individual i.m. (B) or s.c. (D) tumors. Horizontal lines represent mean values.
fractions than the TS-415 tumors, regardless of whether the tumors were transplanted i.m. (p = 0.019) or s.c. (p = 0.039). The hypoxic fractions were higher for the s.c. tumors than for the i.m. tumors, both for the CK-160 (P = 0.010) and the TS-415 (p = 0.0007) line. The four groups of tumors differed in Ktrans as well as in fraction of radiobiologically hypoxic cells. The order of the groups from high to low Ktrans values (i.e., i.m. TS-415, i.m. CK-160, s.c. TS-415, s.c. CK-160) was the same as the order of the groups from low to high hypoxic fractions (i.e., i.m. TS-415, i.m. CK-160, s.c. TS-415, s.c. CK-160). This is illustrated in Fig. 5, which shows a plot of Ktrans vs. fraction of radiobiologically hypoxic cells. On the other hand, there was no relationship between ve and fraction of radiobiologically hypoxic cells.
A noninvasive assay for the extent of hypoxia in squamous cell carcinoma of the uterine cervix is needed, and the potential of DCE-MRI was investigated in the present study. Two human cervical carcinoma xenograft lines differing in differentiation status, cellular radiation sensitivity, and extent of hypoxia were used as tumor models. The poorly differentiated TS-415 tumors were more resistant to radiation than the welldifferentiated CK-160 tumors (i.e., the D0 was higher for TS-415 than for CK-160), whereas the CK-160 tumors were more hypoxic than the TS-415 tumors. By varying the transplantation site, tumors differing widely in extent of hypoxia were obtained, a prerequisite for the present investigation. Tumor hypoxia was assessed by measuring fraction of radiobiologically hypoxic cells using the paired survival curve method. This assay measures the fraction of the clonogenic cells in tumors that are hypoxic during the radiation exposure, and the assay detects both chronically and acutely hypoxic cells. Fraction of radiobiologically hypoxic cells is considered to be of greater clinical significance than hypoxic fractions derived from nonradiobiologic assays, as only clonogenic cells are of relevance for tumor growth and response to treatment (29). Immunohistochemical assays and electrode-based pO2 assays may be useful supplements to radiobiologic assays; but because these assays detect nonclonogenic as well as clonogenic cells, they may give erronous estimates of the extent of hypoxia in tumors (29). The DCE-MRI technique was performed at a time resolution of 14 s and a voxel size of 0.23 0.47 2.0 mm3. Under these conditions, well-defined Gd-DTPA concentration vs. time curves were produced at the single-voxel level, and the signal-to-noise ratio was sufficiently high that the derived parametric images were not influenced significantly by noise, as accounted for earlier (30, 31) and confirmed here. It was also confirmed that tumor motion did not occur during the 15-min acquisition period (i.e., any tumor motion was insignificant relative to the voxel size), as illustrated previously (23). By subjecting the same tumors to DCE-MRI twice, we have shown that highly reproducible parametric images are produced by the DCE-MRI method used here (25, 30). The method of contrast administration may represent a limitation of our study, as Gd-DTPA was administered manually rather than by an infusion pump, and the DCE-MRI data were analyzed with the use of a mean arterial input function rather than individual arterial input functions, thus ignoring any variation in the injection rate among mice (23). Another limitation is related to the tumor model. The volume of the tumors studied here was 200 to 600 mm3, whereas the primary tumors of cervical cancer patients treated with radiation therapy have a volume that is typically multiple times larger (6). The DCE-MRI series were analyzed by using the Tofts generalized pharmacokinetic model (26). Two types of parametric images were derived from the DCE-MRI series: ve images that represent the extracellular volume fraction of the imaged tissue, and Ktrans images that represent the volume
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
843
Fig. 4. Cell surviving fraction vs. radiation dose for intramuscular (i.m.) CK-160 tumors, subcutaneous (s.c.) CK-160 tumors, i.m. TS-415 tumors, and s.c. TS-415 tumors irradiated under clamped (C) or unclamped (B) conditions. Points and bars represent geometric mean SD of five to eight tumors.
transfer constant of the contrast agent. Ktrans equals the blood plasma flow per unit volume of tissue under flow-limited conditions, and the permeability surface area product per unit volume of tissue under permeability-limited conditions (26). For low–molecular weight contrast agents such as Gd-DTPA and tissues with high microvascular permeability such as tumor tissues, Ktrans is believed to be determined primarily by the blood perfusion (11, 26). The Ktrans values differed significantly among the four tumor groups, as did fraction of radiobiologically hypoxic cells, whereas significant differences in ve could not be detected. Interestingly, an unambiguous inverse relationship was found between Ktrans and fraction of radiobiologically hypoxic cells. Because tumor hypoxia is a result of an imbalance between oxygen supply and oxygen consumption (20), the differences in fraction of radiobiologically hypoxic cells were probably a result of differences in blood perfusion rather than differences in cell density, and hence, a consequence of differences in rate of oxygen supply rather than differences in rate of oxygen consumption. This observation suggests that DCE-MRI may provide information on fraction of hypoxic cells in human cervical carcinoma xenografts, and it opens the possibility that DCE-MRI may be developed to be a clinically useful method
for assessing the extent of hypoxia in carcinoma of the cervix, suggestions that are consistent with previous studies of human melanoma xenografts in our laboratory. A strong inverse correlation was found between a DCE-MRI–derived blood perfusion parameter and hypoxic fraction when individual A-07 melanoma xenografts were subjected to DCE-MRI
Fig. 5. Ktrans vs. fraction of radiobiologically hypoxic cells for human cervical carcinoma xenografts. Points and bars represent mean SE, calculated from the data in Fig. 3a and Fig. 4.
844
I. J. Radiation Oncology d Biology d Physics
and measurement of radiobiologic hypoxia (22). However, DCE-MRI studies of human melanoma xenograft lines differing substantially in extracellular volume fraction have suggested that a DCE-MRI-based assay for hypoxia in melanomas should involve DCE-MRI–derived parameters for extracellular volume fraction as well as for blood perfusion (23). This is likely to be the situation also for carcinoma of the cervix, as histologic examinations have shown that the cell density may differ significantly among individual cervical tumors (17, 32). Clinical studies comparing DCE-MRI–derived parameters with tumor oxygenation as measured with Eppendorf pO2 electrodes have been performed in locally advanced cervical carcinoma. Correlations were found between median pO2 or fraction of pO2 readings less than 5 mm Hg on one hand and maximum contrast enhancement over baseline, rate of contrast enhancement, or amplitude of contrast enhancement on the other (17–19). Most of the correlations were on the borderline of being statistically significant, thus questioning the potential of DCE-MRI as a clinically useful method for assessing the extent of hypoxia in cervical carcinoma. However, there are several possible explanations for the weak correlations, including poor temporal and spatial resolution of the DCE-MRI and the use of inadequate DCE-MRI– derived parameters. Thus, the DCE-MRI–derived parameters were based on signal intensities rather than Gd-DTPA concentrations; and they were influenced significantly by
Volume 73, Number 3, 2009
several physiologic parameters of the tumor tissue in an unpredictable way, including blood perfusion, blood volume fraction, microvascular permeability, and extracellular volume fraction. Clinical investigations of the value of DCEMRI as a method for assessing the extent of hypoxia in carcinoma of the cervix may thus require improved techniques of data acquisition and analysis compared with those used thus far. It should be noted, however, that clinical studies investigating the prognostic value of DCE-MRI in locally advanced cervical carcinoma are currently being performed in several cancer centers, and the initial studies have indeed given promising results. Thus it has been reported that high contrast enhancement may be associated with increased probability of local tumor control (13, 14) and disease-specific survival (19) in patients treated with radiation therapy. The study reported here suggests that the rate of contrast enhancement rather than the magnitude of contrast enhancement may be related to the extent of radiobiologic hypoxia, and, hence, the outcome of radiation therapy in cervical carcinoma. In summary, the present preclinical investigation suggests that valuable information on the extent of hypoxia in human cervical carcinoma xenografts can be obtained by use of GdDTPA–based DCE-MRI. This observation opens the possibility that DCE-MRI may be developed as a clinically useful method for assessing tumor oxygenation status in carcinoma of the cervix.
REFERENCES 1. Stehman FB, Perez CA, Kurman RJ, Thigpen JT. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia, PA: Lippincott-Raven Press; 1997. p. 785–857. 2. Kolstad P. Intercapillary distance, oxygen tension and local recurrence in cervix cancer. Scand J Clin Lab Invest 1968; 106:145–157. 3. Ho¨ckel M, Schlenger K, Aral B, et al. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 1996;56:4509–4515. 4. Ho¨ckel M, Knoop C, Schlenger K, et al. Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 1993;26:45–50. 5. Fyles AW, Milosevic M, Wong R, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998;48:149–156. 6. Sundfør K, Lyng H, Trope´ CG, Rofstad EK. Treatment outcome in advanced squamous cell carcinoma of the uterine cervix: Relationships to pretreatment tumor oxygenation and vascularization. Radiother Oncol 2000;54:101–107. 7. Sundfør K, Lyng H, Rofstad EK. Tumour hypoxia and vascular density as predictors of metastasis in squamous cell carcinoma of the uterine cervix. Br J Cancer 1998;78:822–827. 8. Pitson G, Fyles A, Milosevic M, et al. Tumor size and oxygenation are independent predictors of nodal disease in patients with cervix cancer. Int J Radiat Oncol Biol Phys 2001;51: 699–703. 9. Nordsmark M, Loncaster J, Aquino-Parsons C, et al. The prognostic value of pimonidazole and tumour pO2 in human cervix carcinomas after radiation therapy: A prospective international multi-center study. Radiother Oncol 2006;80: 123–131.
10. Fyles A, Milosevic M, Pintilie M, et al. Long-term performance of interstitial fluid pressure and hypoxia as prognostic factors in cervix cancer. Radiother Oncol 2006;80:132–137. 11. Padhani AR. Dynamic contrast-enhanced MRI in clinical oncology: Current status and future directions. J Magn Reson Imaging 2002;16:407–422. 12. Yamashita Y, Takahashi M, Sawada T, et al. Carcinoma of the cervix: Dynamic MR imaging. Radiology 1992;182:643–648. 13. Mayr NA, Yuh WTC, Magnotta VA, et al. Tumor perfusion studies using fast magnetic resonance imaging technique in advanced cervical cancer: A new noninvasive predictive assay. Int J Radiat Oncol Biol Phys 1996;36:623–633. 14. Mayr NA, Yuh WTC, Zheng J, et al. Prediction of tumor control in patients with cervical cancer: Analysis of combined volume and dynamic enhancement pattern by MR imaging. Am J Roentgenol 1998;170:177–182. 15. Hawighorst H, Knapstein PG, Weikel W, et al. Angiogenesis of uterine cervical carcinoma: Characterization by pharmacokinetic magnetic resonance parameters and histological microvessel density with correlation to lymphatic involvement. Cancer Res 1997;57:4777–4786. 16. Yamashita Y, Baba T, Baba Y, et al. Dynamic contrast-enhanced MR imaging of uterine cervical cancer: Pharmacokinetic analysis with histopathologic correlation and its importance in predicting the outcome of radiation therapy. Radiology 2000;216: 803–809. 17. Lyng H, Vorren AO, Sundfør K, et al. Assessment of tumor oxygenation in human cervical carcinoma by use of dynamic Gd-DTPA-enhanced MR imaging. J Magn Reson Imaging 2001;14:750–756. 18. Cooper RA, Carrington BM, Loncaster JA, et al. Tumour oxygenation levels correlate with dynamic contrast-enhanced
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
19.
20. 21. 22.
23.
24. 25.
magnetic resonance imaging parameters in carcinoma of the cervix. Radiother Oncol 2000;57:53–59. Loncaster JA, Carrington BM, Sykes JR, et al. Prediction of radiotherapy outcome using dynamic contrast enhanced MRI of carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2002;54:759–767. Gulledge CJ, Dewhirst MW. Tumor oxygenation: A matter of supply and demand. Anticancer Res 1996;16:741–750. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors. Cancer Res 1989;49:6449–6465. Egeland TAM, Gaustad JV, Vestvik IK, et al. Assessment of fraction of radiobiologically hypoxic cells in human melanoma xenografts by dynamic contrast-enhanced MRI. Magn Reson Med 2006;55:874–882. Vestvik IK, Egeland TAM, Gaustad JV, Mathiesen B, Rofstad EK. Assessment of microvascular density, extracellular volume fraction, and radiobiological hypoxia in human melanoma xenografts by dynamic contrast-enhanced MRI. J Magn Reson Imaging 2007;26:1033–1042. Hittmair K, Gomiscek G, Langenberger K, et al. Method for the quantitative assessment of contrast agent uptake in dynamic contrast-enhanced MRI. Magn Reson Med 1994;31:567–571. Benjaminsen IC, Graff BA, Brurberg KG, Rofstad EK. Assessment of tumor blood perfusion by high-resolution dynamic contrast-enhanced MRI: A preclinical study of human melanoma xenografts. Magn Reson Med 2004;52:269–276.
845
26. Tofts PS, Brix G, Buckley DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusable tracer: Standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–232. 27. Rofstad EK, Ma˚seide K. Radiobiological and immunohistochemical assessment of hypoxia in human melanoma xenografts: Acute and chronic hypoxia in individual tumours. Int J Radiat Biol 1999;75:1377–1393. 28. Rofstad EK. Influence of cellular radiation sensitivity on local tumor control of human melanoma xenografts given fractionated radiation treatment. Cancer Res 1991;51:4609–4612. 29. Fenton BM, Kiani MF, Siemann DW. Should direct measurements of tumor oxygenation relate to the radiobiological hypoxic fraction of a tumor? Int J Radiat Oncol Biol Phys 1995; 33:365–373. 30. Benjaminsen IC, Brurberg KG, Ruud EB, Rofstad EK. Assessment of extravascular extracellular space fraction in human melanoma xenografts by DCE-MRI and kinetic modeling. Magn Reson Imaging 2008;26:160–170. 31. Brurberg KG, Benjaminsen IC, Dørum LMR, Rofstad EK. Fluctuations in tumor blood perfusion assessed by dynamic contrast-enhanced MRI. Magn Reson Med 2007;58:473–481. 32. Lyng H, Sundfør K, Trope´ C, Rofstad EK. Disease control of uterine cervical cancer: Relationships to tumor oxygen tension, vascular density, cell density, and frequency of mitosis and apoptosis measured before treatment and during radiotherapy. Clin Cancer Res 2000;6:1104–1112.
doi:10.1016/j.ijrobp.2008.10.062
BIOLOGY CONTRIBUTION
ASSESSMENT OF HYPOXIA IN HUMAN CERVICAL CARCINOMA XENOGRAFTS BY DYNAMIC CONTRAST-ENHANCED MAGNETIC RESONANCE IMAGING CHRISTINE ELLINGSEN, M.SC., TORMOD A. M. EGELAND, M.SC., KRISTINE GULLIKSRUD, M.SC., JON-VIDAR GAUSTAD, M.SC., BERIT MATHIESEN, B.SC., AND EINAR K. ROFSTAD, PH.D. Group of Radiation Biology and Tumor Physiology, Department of Radiation Biology, Institute for Cancer Research, Norwegian Radium Hospital, Montebello, Oslo, Norway Purpose: Patients with advanced cervical cancer and highly hypoxic primary tumors show increased frequency of locoregional treatment failure and poor disease-free and overall survival rates. The potential usefulness of gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA)–based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) in assessing tumor hypoxia noninvasively was investigated in the present preclinical study. Methods and Materials: CK-160 and TS-415 human cervical carcinoma xenografts transplanted intramuscularly (i.m.) or subcutaneously (s.c.) in BALB/c nu/nu mice were subjected to DCE-MRI and measurement of fraction of radiobiologically hypoxic cells. Tumor images of Ktrans (the volume transfer constant of Gd-DTPA) and ve (the extracellular volume fraction of the imaged tissue) were produced by pharmacokinetic analysis of the DCE-MRI data. Fraction of radiobiologically hypoxic cells was measured by using the paired survival curve method. Results: Fraction of radiobiologically hypoxic cells differed significantly among the four tumor groups. The mean values ± SE were determined to be 44% ± 7% (i.m. CK-160), 77% ± 10% (s.c. CK-160), 23% ± 5% (i.m. TS-415), and 52% ± 6% (s.c. TS-415). The four tumor groups differed significantly also in Ktrans, and there was an unambiguous inverse relationship between Ktrans and fraction of radiobiologically hypoxic cells. On the other hand, significant differences among the groups in ve could not be detected. Conclusions: The study supports the clinical development of DCE-MRI as a method for assessing the extent of hypoxia in carcinoma of the cervix. Ó 2009 Elsevier Inc. Cervical carcinoma, DCE-MRI, Hypoxia, Radiation sensitivity, Xenografts.
Preclinical and clinical studies have suggested that gadolinium-diethylenetriaminepentaacetic acid (Gd-DTPA)–based dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) may be a useful noninvasive method for characterizing the physiologic microenvironment of tumors (11). A DCE-MRI–based method is particularly attractive in cervical carcinoma patients because DCE-MRI is an established and commonly used method for depicting the primary tumor in these patients (12). Moreover, initial investigations of the prognostic value of DCE-MRI in carcinoma of the cervix have given promising results. Mayr et al (13, 14) observed that a high relative signal intensity in the primary tumor was associated with a high local control rate in patients given radiation therapy. High levels of contrast enhancement could be a result of efficient blood perfusion and thus reflect good oxygenation and hence increased radioresponsiveness. Alternatively, high levels of contrast enhancement could be a result of large extracellular volume fractions and thus reflect low numbers of clonogenic cells and hence increased radiocurability.
INTRODUCTION Advanced squamous cell carcinoma of the uterine cervix is primarily treated with radiation alone or radiation in combination with surgery and/or chemotherapy (1). In the 1960s, hypoxia was identified as a characteristic feature of cervical carcinoma that could cause resistance to radiation therapy (2). Recent studies have confirmed these initial observations (3–6) and have also shown that hypoxia may be an adverse prognostic factor in patients given surgery as primary treatment (3). Moreover, extensive hypoxia in the primary tumor has been shown to be associated with biologic aggressiveness, invasive growth, and increased metastatic propensity (7, 8). Thus, cervical carcinoma patients with highly hypoxic tumors show increased frequency of locoregional treatment failure and poor disease-free and overall survival rates (9, 10), and these patients may therefore benefit from particularly aggressive treatment. Consequently, a noninvasive diagnostic method for assessing the extent of hypoxia in cervical tumors is needed.
Acknowledgments—This study was supported by the Norwegian Cancer Society. Received Aug 11, 2008, and in revised form Oct 3, 2008. Accepted for publication Oct 8, 2008.
Reprint requests to: Einar K. Rofstad, Ph.D., Department of Radiation Biology, Institute for Cancer Research, Norwegian Radium Hospital, Montebello, N-0310 Oslo, Norway. Tel: 47-2278-1206; Fax: 47-2278-1207; E-mail: [email protected] Conflict of interest: none. 838
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
There has been an increasing interest in using DCE-MRI for characterizing the physiologic microenvironment of cervical tumors (15–19). Correlations have been found between DCE-MRI–derived parameters and microvascular density, amount of interstitial fibrous tissue, tumor cell density, and extracellular volume fraction as assessed by histologic examinations of the imaged tissue (15–17) as well as tumor oxygenation as assessed by polarographic measurements of tissue pO2 (17–19). However, the correlations are generally weak, perhaps because unsuitable parameters were derived from the DCE-MRI series (i.e., parameters calculated from signal intensity curves without the use of pharmacokinetic models or by the use of inadequate models) or because the DCE-MRI was performed under suboptimal conditions (i.e., inadequate temporal and spatial resolution, low signalto-noise ratio, and/or significant motion artifacts). High-quality preclinical studies are therefore needed to establish whether DCE-MRI has the potential to provide clinically useful surrogate parameters for the extent of hypoxia in cervical carcinoma. The potential usefulness of Gd-DTPA-based DCE-MRI in assessing fraction of hypoxic cells in human tumor xenografts is currently being evaluated in our laboratory. Our studies are based on the hypothesis that tumor hypoxia is the result of an imbalance between oxygen supply and oxygen consumption (20). The oxygen supply is determined primarily by the blood perfusion, and the oxygen consumption is determined primarily by the respiratory activity of the tissue and, hence, the cell density (21). Hypoxic tissue is therefore expected to be found in tumor regions with poor blood perfusion and/or low extracellular volume fraction. We have already shown that DCE-MRI–derived parametric images may provide information on the extent of hypoxia in human melanoma xenografts (22, 23). The purpose of the work reported here was to investigate the feasibility of DCE-MRI in assessing the extent of hypoxia in human cervical carcinoma xenografts. Tumors of two xenograft lines differing significantly in histologic appearance, cellular radiation sensitivity, and fraction of radiobiologically hypoxic cells were included in the study. METHODS AND MATERIALS Mice and tumors CK-160 and TS-415 human cervical carcinoma xenografts growing in adult (8-12 weeks of age) female BALB/c nu/nu mice were used as experimental tumor models. Tumors were initiated from established cell lines cultured in RPMI-1640 (25 mmol/l HEPES and L-glutamine) supplemented with 13% bovine calf serum, 250 mg/l penicillin, and 50 mg/l streptomycin. The CK-160 line was established from a pelvic lymph node metastasis of a 65-year-old woman who had developed a highly invasive, well-differentiated (histologic Grade I), keratinizing primary tumor. The TS-415 line was derived from a pelvic lymph node metastasis of a 45-year-old woman who had developed a highly invasive, poorly differentiated (histologic Grade III), nonkeratinizing primary tumor. The mice were kept under specific pathogen-free conditions and were given sterilized food and tap water ad libitum. Approximately
839
5 105 cells in 10 ml of Hanks’ balanced salt solution were inoculated in the gastrocnemius muscle to produce intramuscular (i.m.) tumors or in the subcutis far back on the flank to produce subcutaneous (s.c.) tumors. Experiments were initiated when the tumors had grown to a volume of 200 to 600 mm3. Tumor volume (V) was calculated as V = p/6 a b2, where a is the longer and b is the shorter of two perpendicular diameters. Both DCE-MRI and tumor irradiation were carried out with anesthetized mice. Fentanyl citrate (Janssen Pharmaceutica, Beerse, Belgium), fluanisone (Janssen Pharmaceutica), and midazolam (Hoffmann-La Roche, Basel, Switzerland) were administered i.p. in doses of 0.63 mg/kg, 20 mg/kg, and 10 mg/kg, respectively. Animal care and experimental procedures were approved by the Institutional Committee on Research Animal Care and were performed in accordance with the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY).
DCE-MRI The DCE-MRI technique was carried out on 15 i.m. CK-160 tumors, 16 s.c. CK-160 tumors, 15 i.m. TS-415 tumors, and 14 s.c. TS-415 tumors by using a procedure described in detail previously (23). Briefly, a 24-G neoflon connected to a syringe by a polyethylene tubing was inserted in the tail vein of tumor-bearing mice, and GdDTPA (Schering, Berlin, Germany) was administered in a bolus dose of 5.0 ml/kg (0.3 mmol/kg) after the mice had been positioned in the magnet. Imaging was performed at a spatial resolution of 0.23 0.47 2.0 mm3 and a time resolution of 14 s by using a 1.5-T whole-body scanner (Signa, General Electric, Milwaukee, WI) and a 13-cm-long, cylindrical, slotted tube resonator transceiver mouse coil with a diameter of 40 mm. Two calibration tubes, one with 0.5 mmol/l Gd-DTPA in 0.9% saline and the other with 0.9% saline only, were placed adjacent to the mice in the coil. The tumors were imaged axially in a single section through the tumor center by using a scan thickness of 2 mm, a number of excitations of 1, an image matrix of 256 64, and a field of view of 6 3 cm2. Two types of spoiled gradient recalled images were recorded: proton density images with repetition time TR = 900 ms, echo time TE = 3.2 ms, and flip angle a = 20 , and T1-weighted images with TR = 200 ms, TE = 3.2 ms, and a = 80 . Two proton density images and three T1-weighted images were acquired before Gd-DTPA was administered, and T1-weighted images were recorded for 15 min after the administration of Gd-DTPA. The Gd-DTPA concentrations were calculated from signal intensities as described by Hittmair et al (24), and the DCE-MRI series were analyzed on a voxel-by-voxel basis by using the arterial input function established by Benjaminsen et al (25) and the pharmacokinetic model developed by Tofts et al (26). Two parameters were determined for each voxel: Ktrans, the volume transfer constant of Gd-DTPA, and ve, the extracellular volume fraction of the imaged tissue. Images of Ktrans and ve were generated by using the SigmaPlot software (SPSS Inc., Chicago, IL).
Fraction of radiobiologically hypoxic cells Intramuscular CK-160 tumors, s.c. CK-160 tumors, i.m. TS-415 tumors, and s.c. TS-415 tumors were irradiated at a dose rate of 5.1 Gy/min by using a Siemens Stabilipan X-ray unit, operated at 220 kV, 19-20 mA, and with 0.5-mm copper filtration (27). Hypoxic tumors were obtained by occluding the blood supply with a clamp 5 min before irradiation. Tumor cell survival was measured in vitro by using a plastic surface colony assay (28). Briefly, the tumors were resected
I. J. Radiation Oncology d Biology d Physics
840
immediately after irradiation, minced in cold phosphate-buffered saline, and treated with an enzyme solution consisting of 0.2% collagenase, 0.05% pronase, and 0.02% DNase (2 h at 37 C). Trypan blue–negative cells were plated in 25-cm2 tissue culture flasks and incubated at 37 C for 14 days. Cells giving rise to colonies >50 cells were scored as clonogenic. The cell surviving fraction of an irradiated tumor was calculated from the plating efficiency of the cells of the tumor and the mean plating efficiency of the cells of six untreated control tumors. Cell survival curves were established for clamped and unclamped tumors, and the fraction of radiobiologically hypoxic cells was calculated from the vertical displacement of the curves (27).
Statistical analysis Experimental data are presented as arithmetic mean SE unless otherwise stated. Statistical comparisons of data were carried out by using the Student t test (single comparisons) or by one-way analysis of variance (multiple comparisons) followed by the Bonferroni test when the data complied with the conditions of normality and equal variance. Under other conditions, comparisons were carried out by nonparametric analysis using the Mann-Whitney rank-sum test (single comparisons) or the Kruskal-Wallis one-way analysis of variance on ranks (multiple comparisons) followed by Dunn’s test. Probability values of p < 0.05, determined from two-sided tests,
Volume 73, Number 3, 2009
were considered significant. The statistical analysis was carried out by using the SigmaStat statistical software (SPSS Inc., Chicago, IL).
RESULTS Histologic findings of the xenografted tumors were similar to those of the donor patients’ tumors. The CK-160 cells gave rise to well-differentiated tumors with a histologic appearance consistent with Grade 1 squamous cell cervical carcinoma (Fig. 1a), whereas the TS-415 cells gave rise to poorly differentiated tumors consistent with Grade 3 squamous cell cervical carcinoma (Fig. 1b). The DCE-MRI data of a representative tumor of each of the four groups are presented in Fig. 2, which shows the Ktrans image, the Ktrans frequency distribution, the ve image, and the ve frequency distribution of an i.m. CK-160 tumor (Fig. 2a), a s.c. CK-160 tumor (Fig. 2b), an i.m. TS-415 tumor (Fig. 2c), and a s.c. TS-415 tumor (Fig. 2d). Significant intratumor heterogeneity in Ktrans was observed, with the highest values in the periphery and the lowest values in central regions. The tumors were also highly heterogeneous in ve and showed regions with low or high values both in the periphery and in the center.
Fig. 1. Histologic appearance of an intramuscular (i.m.) CK-160 tumor (a) and an i.m. TS-415 tumor (b). Sections were stained with hematoxylin.
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
841
Fig. 2. Ktrans image, Ktrans frequency distribution, ve image, and ve frequency distribution of an i.m. CK-160 tumor (a), a s.c. CK-160 tumor (b), an i.m. TS-415 tumor (c), and a s.c. TS-415 tumor (d). (For discussion, see text).
Median Ktrans and median ve of all tumors subjected to DCE-MRI are summarized in Fig. 3. The TS-415 tumors showed higher Ktrans values than the CK-160 tumors, regardless of whether the tumors were transplanted i.m. (p < 0.0001) or s.c. (p = 0.0057). In addition, Ktrans was higher for the i.m. tumors than for the s.c. tumors, both for the CK-160 (p = 0.017) and the TS-415 (p < 0.0001) line. On the other hand, ve did not differ significantly between CK-160 and TS-415 tumors (p > 0.05, both for i.m. and s.c. tumors) or between i.m. and s.c. tumors (p > 0.05, both for CK-160 and TS-415 tumors). Cell survival curves for clamped and unclamped tumors irradiated in vivo and assayed in vitro are presented in Fig. 4. D0 was higher for the TS-415 tumors than for the
CK-160 tumors by a factor of 1.5 (p < 0.0001), but did not differ between i.m. and s.c. (p > 0.05) tumors or between clamped and unclamped (p > 0.05) tumors of the same line. The numerical values of D0 (mean SE) were 3.02 0.09 Gy (i.m. CK-160), 2.97 0.12 Gy (s.c. CK160), 4.50 0.11 Gy (i.m. TS-415), and 4.45 0.10 Gy (s.c. TS-415) for clamped tumors and 2.95 0.13 Gy (i.m. CK-160), 3.04 0.12 Gy (s.c. CK-160), 4.54 0.14 Gy (i.m. TS-415), and 4.49 0.14 Gy (s.c. TS-415) for unclamped tumors. Fraction of radiobiologically hypoxic cells (mean SE) was calculated from these survival curves to be 44% 7% (i.m. CK-160), 77% 10% (s.c. CK-160), 23% 5% (i.m. TS-415), and 52% 6% (s.c. TS-415). The CK-160 tumors showed higher hypoxic
842
I. J. Radiation Oncology d Biology d Physics
Volume 73, Number 3, 2009
DISCUSSION
Fig. 3. Median Ktrans (a) and median ve (b) of 15 intramuscular (i.m.) CK-160 tumors, 16 subcutaneous (s.c.) CK-160 tumors, 15 i.m. TS-415 tumors, and 14 s.c. TS-415 tumors (for discussion, see text). Symbols represent individual i.m. (B) or s.c. (D) tumors. Horizontal lines represent mean values.
fractions than the TS-415 tumors, regardless of whether the tumors were transplanted i.m. (p = 0.019) or s.c. (p = 0.039). The hypoxic fractions were higher for the s.c. tumors than for the i.m. tumors, both for the CK-160 (P = 0.010) and the TS-415 (p = 0.0007) line. The four groups of tumors differed in Ktrans as well as in fraction of radiobiologically hypoxic cells. The order of the groups from high to low Ktrans values (i.e., i.m. TS-415, i.m. CK-160, s.c. TS-415, s.c. CK-160) was the same as the order of the groups from low to high hypoxic fractions (i.e., i.m. TS-415, i.m. CK-160, s.c. TS-415, s.c. CK-160). This is illustrated in Fig. 5, which shows a plot of Ktrans vs. fraction of radiobiologically hypoxic cells. On the other hand, there was no relationship between ve and fraction of radiobiologically hypoxic cells.
A noninvasive assay for the extent of hypoxia in squamous cell carcinoma of the uterine cervix is needed, and the potential of DCE-MRI was investigated in the present study. Two human cervical carcinoma xenograft lines differing in differentiation status, cellular radiation sensitivity, and extent of hypoxia were used as tumor models. The poorly differentiated TS-415 tumors were more resistant to radiation than the welldifferentiated CK-160 tumors (i.e., the D0 was higher for TS-415 than for CK-160), whereas the CK-160 tumors were more hypoxic than the TS-415 tumors. By varying the transplantation site, tumors differing widely in extent of hypoxia were obtained, a prerequisite for the present investigation. Tumor hypoxia was assessed by measuring fraction of radiobiologically hypoxic cells using the paired survival curve method. This assay measures the fraction of the clonogenic cells in tumors that are hypoxic during the radiation exposure, and the assay detects both chronically and acutely hypoxic cells. Fraction of radiobiologically hypoxic cells is considered to be of greater clinical significance than hypoxic fractions derived from nonradiobiologic assays, as only clonogenic cells are of relevance for tumor growth and response to treatment (29). Immunohistochemical assays and electrode-based pO2 assays may be useful supplements to radiobiologic assays; but because these assays detect nonclonogenic as well as clonogenic cells, they may give erronous estimates of the extent of hypoxia in tumors (29). The DCE-MRI technique was performed at a time resolution of 14 s and a voxel size of 0.23 0.47 2.0 mm3. Under these conditions, well-defined Gd-DTPA concentration vs. time curves were produced at the single-voxel level, and the signal-to-noise ratio was sufficiently high that the derived parametric images were not influenced significantly by noise, as accounted for earlier (30, 31) and confirmed here. It was also confirmed that tumor motion did not occur during the 15-min acquisition period (i.e., any tumor motion was insignificant relative to the voxel size), as illustrated previously (23). By subjecting the same tumors to DCE-MRI twice, we have shown that highly reproducible parametric images are produced by the DCE-MRI method used here (25, 30). The method of contrast administration may represent a limitation of our study, as Gd-DTPA was administered manually rather than by an infusion pump, and the DCE-MRI data were analyzed with the use of a mean arterial input function rather than individual arterial input functions, thus ignoring any variation in the injection rate among mice (23). Another limitation is related to the tumor model. The volume of the tumors studied here was 200 to 600 mm3, whereas the primary tumors of cervical cancer patients treated with radiation therapy have a volume that is typically multiple times larger (6). The DCE-MRI series were analyzed by using the Tofts generalized pharmacokinetic model (26). Two types of parametric images were derived from the DCE-MRI series: ve images that represent the extracellular volume fraction of the imaged tissue, and Ktrans images that represent the volume
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
843
Fig. 4. Cell surviving fraction vs. radiation dose for intramuscular (i.m.) CK-160 tumors, subcutaneous (s.c.) CK-160 tumors, i.m. TS-415 tumors, and s.c. TS-415 tumors irradiated under clamped (C) or unclamped (B) conditions. Points and bars represent geometric mean SD of five to eight tumors.
transfer constant of the contrast agent. Ktrans equals the blood plasma flow per unit volume of tissue under flow-limited conditions, and the permeability surface area product per unit volume of tissue under permeability-limited conditions (26). For low–molecular weight contrast agents such as Gd-DTPA and tissues with high microvascular permeability such as tumor tissues, Ktrans is believed to be determined primarily by the blood perfusion (11, 26). The Ktrans values differed significantly among the four tumor groups, as did fraction of radiobiologically hypoxic cells, whereas significant differences in ve could not be detected. Interestingly, an unambiguous inverse relationship was found between Ktrans and fraction of radiobiologically hypoxic cells. Because tumor hypoxia is a result of an imbalance between oxygen supply and oxygen consumption (20), the differences in fraction of radiobiologically hypoxic cells were probably a result of differences in blood perfusion rather than differences in cell density, and hence, a consequence of differences in rate of oxygen supply rather than differences in rate of oxygen consumption. This observation suggests that DCE-MRI may provide information on fraction of hypoxic cells in human cervical carcinoma xenografts, and it opens the possibility that DCE-MRI may be developed to be a clinically useful method
for assessing the extent of hypoxia in carcinoma of the cervix, suggestions that are consistent with previous studies of human melanoma xenografts in our laboratory. A strong inverse correlation was found between a DCE-MRI–derived blood perfusion parameter and hypoxic fraction when individual A-07 melanoma xenografts were subjected to DCE-MRI
Fig. 5. Ktrans vs. fraction of radiobiologically hypoxic cells for human cervical carcinoma xenografts. Points and bars represent mean SE, calculated from the data in Fig. 3a and Fig. 4.
844
I. J. Radiation Oncology d Biology d Physics
and measurement of radiobiologic hypoxia (22). However, DCE-MRI studies of human melanoma xenograft lines differing substantially in extracellular volume fraction have suggested that a DCE-MRI-based assay for hypoxia in melanomas should involve DCE-MRI–derived parameters for extracellular volume fraction as well as for blood perfusion (23). This is likely to be the situation also for carcinoma of the cervix, as histologic examinations have shown that the cell density may differ significantly among individual cervical tumors (17, 32). Clinical studies comparing DCE-MRI–derived parameters with tumor oxygenation as measured with Eppendorf pO2 electrodes have been performed in locally advanced cervical carcinoma. Correlations were found between median pO2 or fraction of pO2 readings less than 5 mm Hg on one hand and maximum contrast enhancement over baseline, rate of contrast enhancement, or amplitude of contrast enhancement on the other (17–19). Most of the correlations were on the borderline of being statistically significant, thus questioning the potential of DCE-MRI as a clinically useful method for assessing the extent of hypoxia in cervical carcinoma. However, there are several possible explanations for the weak correlations, including poor temporal and spatial resolution of the DCE-MRI and the use of inadequate DCE-MRI– derived parameters. Thus, the DCE-MRI–derived parameters were based on signal intensities rather than Gd-DTPA concentrations; and they were influenced significantly by
Volume 73, Number 3, 2009
several physiologic parameters of the tumor tissue in an unpredictable way, including blood perfusion, blood volume fraction, microvascular permeability, and extracellular volume fraction. Clinical investigations of the value of DCEMRI as a method for assessing the extent of hypoxia in carcinoma of the cervix may thus require improved techniques of data acquisition and analysis compared with those used thus far. It should be noted, however, that clinical studies investigating the prognostic value of DCE-MRI in locally advanced cervical carcinoma are currently being performed in several cancer centers, and the initial studies have indeed given promising results. Thus it has been reported that high contrast enhancement may be associated with increased probability of local tumor control (13, 14) and disease-specific survival (19) in patients treated with radiation therapy. The study reported here suggests that the rate of contrast enhancement rather than the magnitude of contrast enhancement may be related to the extent of radiobiologic hypoxia, and, hence, the outcome of radiation therapy in cervical carcinoma. In summary, the present preclinical investigation suggests that valuable information on the extent of hypoxia in human cervical carcinoma xenografts can be obtained by use of GdDTPA–based DCE-MRI. This observation opens the possibility that DCE-MRI may be developed as a clinically useful method for assessing tumor oxygenation status in carcinoma of the cervix.
REFERENCES 1. Stehman FB, Perez CA, Kurman RJ, Thigpen JT. Uterine cervix. In: Hoskins WJ, Perez CA, Young RC, editors. Principles and practice of gynecologic oncology. 2nd ed. Philadelphia, PA: Lippincott-Raven Press; 1997. p. 785–857. 2. Kolstad P. Intercapillary distance, oxygen tension and local recurrence in cervix cancer. Scand J Clin Lab Invest 1968; 106:145–157. 3. Ho¨ckel M, Schlenger K, Aral B, et al. Association between tumor hypoxia and malignant progression in advanced cancer of the uterine cervix. Cancer Res 1996;56:4509–4515. 4. Ho¨ckel M, Knoop C, Schlenger K, et al. Intratumoral pO2 predicts survival in advanced cancer of the uterine cervix. Radiother Oncol 1993;26:45–50. 5. Fyles AW, Milosevic M, Wong R, et al. Oxygenation predicts radiation response and survival in patients with cervix cancer. Radiother Oncol 1998;48:149–156. 6. Sundfør K, Lyng H, Trope´ CG, Rofstad EK. Treatment outcome in advanced squamous cell carcinoma of the uterine cervix: Relationships to pretreatment tumor oxygenation and vascularization. Radiother Oncol 2000;54:101–107. 7. Sundfør K, Lyng H, Rofstad EK. Tumour hypoxia and vascular density as predictors of metastasis in squamous cell carcinoma of the uterine cervix. Br J Cancer 1998;78:822–827. 8. Pitson G, Fyles A, Milosevic M, et al. Tumor size and oxygenation are independent predictors of nodal disease in patients with cervix cancer. Int J Radiat Oncol Biol Phys 2001;51: 699–703. 9. Nordsmark M, Loncaster J, Aquino-Parsons C, et al. The prognostic value of pimonidazole and tumour pO2 in human cervix carcinomas after radiation therapy: A prospective international multi-center study. Radiother Oncol 2006;80: 123–131.
10. Fyles A, Milosevic M, Pintilie M, et al. Long-term performance of interstitial fluid pressure and hypoxia as prognostic factors in cervix cancer. Radiother Oncol 2006;80:132–137. 11. Padhani AR. Dynamic contrast-enhanced MRI in clinical oncology: Current status and future directions. J Magn Reson Imaging 2002;16:407–422. 12. Yamashita Y, Takahashi M, Sawada T, et al. Carcinoma of the cervix: Dynamic MR imaging. Radiology 1992;182:643–648. 13. Mayr NA, Yuh WTC, Magnotta VA, et al. Tumor perfusion studies using fast magnetic resonance imaging technique in advanced cervical cancer: A new noninvasive predictive assay. Int J Radiat Oncol Biol Phys 1996;36:623–633. 14. Mayr NA, Yuh WTC, Zheng J, et al. Prediction of tumor control in patients with cervical cancer: Analysis of combined volume and dynamic enhancement pattern by MR imaging. Am J Roentgenol 1998;170:177–182. 15. Hawighorst H, Knapstein PG, Weikel W, et al. Angiogenesis of uterine cervical carcinoma: Characterization by pharmacokinetic magnetic resonance parameters and histological microvessel density with correlation to lymphatic involvement. Cancer Res 1997;57:4777–4786. 16. Yamashita Y, Baba T, Baba Y, et al. Dynamic contrast-enhanced MR imaging of uterine cervical cancer: Pharmacokinetic analysis with histopathologic correlation and its importance in predicting the outcome of radiation therapy. Radiology 2000;216: 803–809. 17. Lyng H, Vorren AO, Sundfør K, et al. Assessment of tumor oxygenation in human cervical carcinoma by use of dynamic Gd-DTPA-enhanced MR imaging. J Magn Reson Imaging 2001;14:750–756. 18. Cooper RA, Carrington BM, Loncaster JA, et al. Tumour oxygenation levels correlate with dynamic contrast-enhanced
Dynamic contrast-enhanced MRI and tumor hypoxia d C. ELLINGSEN et al.
19.
20. 21. 22.
23.
24. 25.
magnetic resonance imaging parameters in carcinoma of the cervix. Radiother Oncol 2000;57:53–59. Loncaster JA, Carrington BM, Sykes JR, et al. Prediction of radiotherapy outcome using dynamic contrast enhanced MRI of carcinoma of the cervix. Int J Radiat Oncol Biol Phys 2002;54:759–767. Gulledge CJ, Dewhirst MW. Tumor oxygenation: A matter of supply and demand. Anticancer Res 1996;16:741–750. Vaupel P, Kallinowski F, Okunieff P. Blood flow, oxygen and nutrient supply, and metabolic microenvironment of human tumors. Cancer Res 1989;49:6449–6465. Egeland TAM, Gaustad JV, Vestvik IK, et al. Assessment of fraction of radiobiologically hypoxic cells in human melanoma xenografts by dynamic contrast-enhanced MRI. Magn Reson Med 2006;55:874–882. Vestvik IK, Egeland TAM, Gaustad JV, Mathiesen B, Rofstad EK. Assessment of microvascular density, extracellular volume fraction, and radiobiological hypoxia in human melanoma xenografts by dynamic contrast-enhanced MRI. J Magn Reson Imaging 2007;26:1033–1042. Hittmair K, Gomiscek G, Langenberger K, et al. Method for the quantitative assessment of contrast agent uptake in dynamic contrast-enhanced MRI. Magn Reson Med 1994;31:567–571. Benjaminsen IC, Graff BA, Brurberg KG, Rofstad EK. Assessment of tumor blood perfusion by high-resolution dynamic contrast-enhanced MRI: A preclinical study of human melanoma xenografts. Magn Reson Med 2004;52:269–276.
845
26. Tofts PS, Brix G, Buckley DL, et al. Estimating kinetic parameters from dynamic contrast-enhanced T1-weighted MRI of a diffusable tracer: Standardized quantities and symbols. J Magn Reson Imaging 1999;10:223–232. 27. Rofstad EK, Ma˚seide K. Radiobiological and immunohistochemical assessment of hypoxia in human melanoma xenografts: Acute and chronic hypoxia in individual tumours. Int J Radiat Biol 1999;75:1377–1393. 28. Rofstad EK. Influence of cellular radiation sensitivity on local tumor control of human melanoma xenografts given fractionated radiation treatment. Cancer Res 1991;51:4609–4612. 29. Fenton BM, Kiani MF, Siemann DW. Should direct measurements of tumor oxygenation relate to the radiobiological hypoxic fraction of a tumor? Int J Radiat Oncol Biol Phys 1995; 33:365–373. 30. Benjaminsen IC, Brurberg KG, Ruud EB, Rofstad EK. Assessment of extravascular extracellular space fraction in human melanoma xenografts by DCE-MRI and kinetic modeling. Magn Reson Imaging 2008;26:160–170. 31. Brurberg KG, Benjaminsen IC, Dørum LMR, Rofstad EK. Fluctuations in tumor blood perfusion assessed by dynamic contrast-enhanced MRI. Magn Reson Med 2007;58:473–481. 32. Lyng H, Sundfør K, Trope´ C, Rofstad EK. Disease control of uterine cervical cancer: Relationships to tumor oxygen tension, vascular density, cell density, and frequency of mitosis and apoptosis measured before treatment and during radiotherapy. Clin Cancer Res 2000;6:1104–1112.